© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 96345
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ÑĢай
    -0.26
    æľĢé«ĺçļĦ
    -0.25
    å¼±
    -0.25
    emente
    -0.25
    éĢĤå®ľ
    -0.24
    ä¸įéĢĤåIJĪ
    -0.24
    å¾Īä½İ
    -0.24
    çŃīåĽłç´ł
    -0.24
    éĿłè¿ij
    -0.24
    åıªèĥ½è¯´
    -0.24
    POSITIVE LOGITS
    æĶ¹éĿ©åıijå±ķ
    0.26
    æ£ĵ
    0.25
    tridge
    0.24
    inputEmail
    0.24
    ubb
    0.24
    {}]
    0.24
    èĥł
    0.23
    æ²ī浸
    0.23
    (Art
    0.23
    ãĥ£
    0.23
    Activations Density 0.789%

    No Known Activations

    This feature has no known activations.